| *** thorst has joined #openstack-ansible | 00:06 | |
| *** thorst has quit IRC | 00:13 | |
| *** rk4n has joined #openstack-ansible | 00:37 | |
| *** woodard has quit IRC | 00:39 | |
| *** rk4n has quit IRC | 00:41 | |
| *** weezS has joined #openstack-ansible | 00:52 | |
| *** weezS has quit IRC | 00:55 | |
| *** thorst has joined #openstack-ansible | 01:11 | |
| *** thorst has quit IRC | 01:19 | |
| *** npawelek has quit IRC | 01:42 | |
| *** npawelek has joined #openstack-ansible | 01:45 | |
| *** npawelek_ has joined #openstack-ansible | 01:48 | |
| *** thorst has joined #openstack-ansible | 01:49 | |
| *** thetrav has joined #openstack-ansible | 01:58 | |
| *** spotz_zzz is now known as spotz | 01:58 | |
| *** hblah has joined #openstack-ansible | 01:58 | |
| thetrav | openstack_user_config.yml <- external_lb_vip_address | 01:58 |
|---|---|---|
| thetrav | can it be a fqdn? | 01:58 |
| thetrav | more specifically | 01:58 |
| thetrav | I have a fqdn | 01:58 |
| thetrav | I want to know where to shove it | 01:58 |
| thetrav | so that horizon and importantly, the keystone public endpoints, are set to that value instead of an ip (which is sometimes natted) | 01:59 |
| thetrav | nat-ed? | 01:59 |
| logan- | Yes it can be a fqdn but it must be resolvable by the infrastructure as haproxy specifically will not start unless it can resolve the host so it knows which IP to bind | 02:00 |
| *** jamielennox is now known as jamielennox|away | 02:00 | |
| *** jamielennox|away is now known as jamielennox | 02:04 | |
| thetrav | rad | 02:04 |
| *** npawelek has quit IRC | 02:05 | |
| *** npawelek_ is now known as npawelek | 02:05 | |
| *** npawelek has quit IRC | 02:15 | |
| *** npawelek has joined #openstack-ansible | 02:18 | |
| csmart | mrda: \o | 02:31 |
| mrda | csmart: o/ | 02:37 |
| *** rk4n has joined #openstack-ansible | 02:38 | |
| *** rk4n has quit IRC | 02:43 | |
| *** llu has joined #openstack-ansible | 02:44 | |
| *** llu has left #openstack-ansible | 02:44 | |
| *** thorst has quit IRC | 02:46 | |
| *** thorst has joined #openstack-ansible | 02:47 | |
| *** spotz is now known as spotz_zzz | 02:48 | |
| *** Jeffrey4l has quit IRC | 02:52 | |
| *** thorst has quit IRC | 02:56 | |
| *** hblah has quit IRC | 02:56 | |
| *** Jeffrey4l has joined #openstack-ansible | 02:58 | |
| *** thorst has joined #openstack-ansible | 03:28 | |
| *** Mudpuppy_afk has joined #openstack-ansible | 03:35 | |
| *** Mudpuppy_afk has quit IRC | 03:39 | |
| *** thorst has quit IRC | 03:52 | |
| *** thetrav has quit IRC | 04:07 | |
| *** janki has joined #openstack-ansible | 04:13 | |
| *** gouthamr has joined #openstack-ansible | 04:19 | |
| *** thorst has joined #openstack-ansible | 04:34 | |
| *** thorst has quit IRC | 04:39 | |
| *** rk4n has joined #openstack-ansible | 04:40 | |
| *** rk4n has quit IRC | 04:44 | |
| *** thorst has joined #openstack-ansible | 04:46 | |
| *** thorst has quit IRC | 04:50 | |
| *** Jeffrey4l has quit IRC | 05:00 | |
| *** thorst has joined #openstack-ansible | 05:00 | |
| *** Jeffrey4l has joined #openstack-ansible | 05:00 | |
| *** AnarchyAo has quit IRC | 05:14 | |
| *** admin0 has joined #openstack-ansible | 05:23 | |
| *** admin0 has quit IRC | 05:26 | |
| *** admin0 has joined #openstack-ansible | 05:40 | |
| *** admin0 has quit IRC | 05:44 | |
| *** thorst has quit IRC | 06:01 | |
| *** thorst has joined #openstack-ansible | 06:02 | |
| *** thorst has quit IRC | 06:05 | |
| *** thorst has joined #openstack-ansible | 06:05 | |
| *** rk4n has joined #openstack-ansible | 06:09 | |
| *** rk4n has quit IRC | 06:14 | |
| *** rk4n has joined #openstack-ansible | 06:40 | |
| *** rk4n has quit IRC | 06:45 | |
| *** pcaruana has joined #openstack-ansible | 06:49 | |
| *** thorst has quit IRC | 06:49 | |
| *** mgeezy has joined #openstack-ansible | 06:49 | |
| mgeezy | anyone familiar with giving openstack vms external access | 06:50 |
| *** thorst has joined #openstack-ansible | 06:50 | |
| *** thorst has quit IRC | 06:55 | |
| *** admin0 has joined #openstack-ansible | 07:01 | |
| *** gouthamr has quit IRC | 07:01 | |
| *** gouthamr has joined #openstack-ansible | 07:02 | |
| *** thorst has joined #openstack-ansible | 07:02 | |
| *** gouthamr has quit IRC | 07:03 | |
| *** thorst has quit IRC | 07:07 | |
| cloudnull | mgeezy: on my way to sleep, but yes. | 07:12 |
| cloudnull | did you have specific questions? | 07:12 |
| cloudnull | did you need ingress or egress? | 07:12 |
| cloudnull | or both ? | 07:12 |
| mgeezy | both | 07:13 |
| mgeezy | i had spoken to james denton | 07:13 |
| mgeezy | he said there was a preferred way to do it | 07:13 |
| mgeezy | but we disconnected before he was able to elaborate | 07:13 |
| cloudnull | we use provider networks | 07:15 |
| *** berendt has joined #openstack-ansible | 07:15 | |
| *** berendt has quit IRC | 07:15 | |
| cloudnull | vlan tagged network given directly to neutron | 07:15 |
| *** berendt has joined #openstack-ansible | 07:16 | |
| cloudnull | which can either be shared for direct attachment | 07:16 |
| cloudnull | or you can set it up on a router for use with a floating IP | 07:16 |
| cloudnull | for speed I perfer provider networks that are shared. | 07:16 |
| mgeezy | so its just create a shared external network | 07:16 |
| cloudnull | but if IP space is at a premium the quota capabilities of floating ips is nice | 07:16 |
| cloudnull | mgeezy: yup | 07:17 |
| mgeezy | and put on a router? | 07:17 |
| cloudnull | yes if you want a floa t | 07:17 |
| mgeezy | if im behind a switch do i have to do any configuration of the switch to tell it to route ? | 07:17 |
| cloudnull | i don't believe so. | 07:18 |
| mgeezy | ok i will try that again | 07:18 |
| mgeezy | been having a pain trying to get this to work | 07:19 |
| cloudnull | i mean you switch has to be confgured to deal with the tagged traffic | 07:19 |
| cloudnull | but nothing special | 07:19 |
| mgeezy | can you elaborate just a little | 07:20 |
| mgeezy | just a general resource to read up on it | 07:20 |
| mgeezy | networking is not my specialty | 07:20 |
| cloudnull | mgeezy: http://cdn.pasteraw.com/fle9wj80pnxorakl1azafw2n627dlnm | 07:22 |
| mgeezy | thank you! | 07:22 |
| cloudnull | that's what's running in the OSIC cloud | 07:22 |
| cloudnull | mgeezy: some folks do a 1 to 1 NAT others connect the public CIDR directly on to the Aggregation switch. its totally up to you. | 07:25 |
| cloudnull | that said. jamesdenton should be around later. | 07:26 |
| mgeezy | very much appreciate it. i will talk to him as well | 07:26 |
| mgeezy | he was going to put me in contact with some people for openstack support | 07:26 |
| mgeezy | happy to connect with you if you can do the same | 07:26 |
| cloudnull | he may be able to elaborate on what you we're talking about before, (idk the context) | 07:27 |
| mgeezy | no worries | 07:27 |
| cloudnull | evrardjp: may be able to give you some pointers too. | 07:27 |
| cloudnull | when he's online. | 07:27 |
| cloudnull | anyway, i'm off to bed. | 07:27 |
| cloudnull | cheers. | 07:27 |
| cloudnull | best of lucj | 07:27 |
| cloudnull | *luck | 07:27 |
| mgeezy | thanks! | 07:28 |
| *** bsv has joined #openstack-ansible | 07:39 | |
| *** Oku_OS-away is now known as Oku_OS | 07:40 | |
| *** drifterza has joined #openstack-ansible | 07:42 | |
| drifterza | lo | 07:45 |
| *** thorst has joined #openstack-ansible | 07:49 | |
| *** fxpester has joined #openstack-ansible | 07:53 | |
| *** thorst has quit IRC | 07:53 | |
| *** thorst has joined #openstack-ansible | 07:55 | |
| *** thorst has quit IRC | 08:00 | |
| evrardjp | have a good night cloudnull and see you in a few hours I guess :p | 08:00 |
| evrardjp | and hello everyone :) | 08:01 |
| *** thorst has joined #openstack-ansible | 08:02 | |
| *** asettle has joined #openstack-ansible | 08:02 | |
| *** thorst has quit IRC | 08:03 | |
| *** thorst has joined #openstack-ansible | 08:03 | |
| *** karimb has joined #openstack-ansible | 08:04 | |
| *** neilus has joined #openstack-ansible | 08:10 | |
| *** rgogunskiy has joined #openstack-ansible | 08:15 | |
| *** neilus has quit IRC | 08:16 | |
| *** neilus_ has joined #openstack-ansible | 08:17 | |
| *** mgeezy has quit IRC | 08:24 | |
| *** thorst has quit IRC | 08:28 | |
| *** thorst has joined #openstack-ansible | 08:32 | |
| *** neilus_1 has joined #openstack-ansible | 08:37 | |
| *** neilus_ has quit IRC | 08:40 | |
| *** karimb has quit IRC | 08:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Update paste, policy and rootwrap configurations 2016-09-12 https://review.openstack.org/368638 | 08:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Update paste, policy and rootwrap configurations 2016-09-08 https://review.openstack.org/367521 | 08:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Newton 2016-09-12 https://review.openstack.org/367531 | 08:44 |
| *** karimb has joined #openstack-ansible | 08:45 | |
| *** rk4n has joined #openstack-ansible | 08:47 | |
| *** neilus_1 has quit IRC | 08:51 | |
| *** rk4n has quit IRC | 08:53 | |
| asettle | odyssey4me: do we have test info to send out to our volunteers today? | 08:56 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible: Properly namespace cinder_storage_address https://review.openstack.org/368653 | 08:57 |
| *** electrofelix has joined #openstack-ansible | 08:59 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_cinder: Correctly namespace the storage_address variable https://review.openstack.org/368654 | 09:00 |
| *** openstackgerrit has quit IRC | 09:04 | |
| *** openstackgerrit has joined #openstack-ansible | 09:05 | |
| *** neilus_ has joined #openstack-ansible | 09:07 | |
| *** rk4n has joined #openstack-ansible | 09:12 | |
| *** rk4n has quit IRC | 09:17 | |
| *** rk4n has joined #openstack-ansible | 09:18 | |
| *** neilus_1 has joined #openstack-ansible | 09:21 | |
| odyssey4me | asettle for those who don't have the know how to put together their own lab or networking setup, this can be used: https://gist.github.com/odyssey4me/863e84a0f6271712c6d48980cffb958d | 09:21 |
| asettle | odyssey4me: amazing, thank you | 09:21 |
| odyssey4me | I have yet to complete an entire run through end to end myself, but that is what this week is for | 09:21 |
| *** rk4n has quit IRC | 09:24 | |
| *** neilus_ has quit IRC | 09:24 | |
| asettle | odyssey4me: amen :) okay, email out! Thank you | 09:25 |
| *** rk4n has joined #openstack-ansible | 09:27 | |
| *** Mudpuppy_afk has joined #openstack-ansible | 09:37 | |
| *** vnogin has joined #openstack-ansible | 09:41 | |
| *** karimb has quit IRC | 09:41 | |
| *** Mudpuppy_afk has quit IRC | 09:42 | |
| *** dalees has joined #openstack-ansible | 09:43 | |
| *** karimb has joined #openstack-ansible | 09:50 | |
| *** fishcried has joined #openstack-ansible | 09:51 | |
| *** thorst has quit IRC | 10:02 | |
| *** rk4n has quit IRC | 10:04 | |
| *** rk4n has joined #openstack-ansible | 10:05 | |
| *** karimb has quit IRC | 10:07 | |
| *** thorst_ has joined #openstack-ansible | 10:11 | |
| *** rk4n has quit IRC | 10:12 | |
| *** neilus_1 has quit IRC | 10:13 | |
| *** neilus_ has joined #openstack-ansible | 10:13 | |
| *** thorst_ has quit IRC | 10:15 | |
| *** thorst has joined #openstack-ansible | 10:17 | |
| *** thorst has quit IRC | 10:21 | |
| *** markvoelker has joined #openstack-ansible | 10:22 | |
| *** thorst has joined #openstack-ansible | 10:23 | |
| *** markvoelker has quit IRC | 10:26 | |
| *** thorst has quit IRC | 10:27 | |
| *** neilus_ has quit IRC | 10:30 | |
| *** rk4n has joined #openstack-ansible | 10:31 | |
| *** higster has joined #openstack-ansible | 10:32 | |
| *** Fenuks has joined #openstack-ansible | 10:35 | |
| *** hblah has joined #openstack-ansible | 10:37 | |
| *** rk4n has quit IRC | 10:37 | |
| *** rk4n has joined #openstack-ansible | 10:44 | |
| *** rk4n has quit IRC | 10:44 | |
| *** neilus_ has joined #openstack-ansible | 10:46 | |
| *** karimb has joined #openstack-ansible | 10:49 | |
| *** smatzek has joined #openstack-ansible | 10:53 | |
| Fenuks | Hello. Anyone available to answer a couple questions regarding access control to swift? I need a user that can only read-write into swift container, not control entire tenant | 10:53 |
| *** hblah has quit IRC | 10:56 | |
| *** rk4n has joined #openstack-ansible | 11:01 | |
| *** higster has quit IRC | 11:11 | |
| *** higster has joined #openstack-ansible | 11:11 | |
| *** higster_ has joined #openstack-ansible | 11:16 | |
| odyssey4me | urgh, something's going wrong with the rabbitmq connectivity: http://logs.openstack.org/86/367886/1/check/gate-openstack-ansible-os_nova-ansible-func-ubuntu-trusty/bec12f3/logs/neutron/neutron-linuxbridge-agent.log.txt.gz | 11:16 |
| *** higster has quit IRC | 11:19 | |
| *** avlis has joined #openstack-ansible | 11:21 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add functional testing for Calico https://review.openstack.org/368287 | 11:22 |
| *** jwitko has quit IRC | 11:26 | |
| *** thorst has joined #openstack-ansible | 11:29 | |
| *** thorst has quit IRC | 11:32 | |
| *** retreved has joined #openstack-ansible | 11:36 | |
| *** thorst has joined #openstack-ansible | 11:38 | |
| *** thorst has quit IRC | 11:43 | |
| *** rk4n has quit IRC | 11:43 | |
| *** karimb has quit IRC | 11:50 | |
| *** thorst has joined #openstack-ansible | 11:56 | |
| *** thorst has quit IRC | 12:00 | |
| *** markvoelker has joined #openstack-ansible | 12:02 | |
| mhayden | happy monday, folks | 12:04 |
| drifterza | hi | 12:05 |
| *** rk4n has joined #openstack-ansible | 12:05 | |
| automagically | Morning all | 12:08 |
| vnogin | morning osa ) | 12:08 |
| *** bsv has quit IRC | 12:10 | |
| mgariepy | good morning everyone | 12:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 12:14 |
| *** rk4n has quit IRC | 12:16 | |
| *** rk4n has joined #openstack-ansible | 12:19 | |
| *** thorst has joined #openstack-ansible | 12:20 | |
| *** hblah has joined #openstack-ansible | 12:21 | |
| *** thorst has quit IRC | 12:21 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure file modes are 644 for inventory and group_vars for Magnum https://review.openstack.org/367164 | 12:26 |
| *** karimb has joined #openstack-ansible | 12:26 | |
| *** thorst has joined #openstack-ansible | 12:27 | |
| *** hblah has quit IRC | 12:29 | |
| *** pester has joined #openstack-ansible | 12:30 | |
| *** DanyC has joined #openstack-ansible | 12:31 | |
| *** thorst has quit IRC | 12:32 | |
| *** fxpester has quit IRC | 12:32 | |
| Matias | morning | 12:34 |
| Matias | stevelle: are you online? | 12:34 |
| *** thorst has joined #openstack-ansible | 12:36 | |
| *** janki has quit IRC | 12:41 | |
| *** wadeholler has quit IRC | 12:43 | |
| Matias | stevelle: it's the ceilometer_collector_container which is supposed to connect to gnocchi in order to send the samples, isn't it? | 12:46 |
| Matias | stevelle: I can't seem to reproduce the setup reliably | 12:47 |
| Matias | I destroyed/recreated the ceilometer and gnocchi containers multiple times | 12:47 |
| Matias | in one of these times, it worked, but a lot of samples were in queue (on rabbit), so gnocchi would take very long to start to consume fresh samples | 12:48 |
| Matias | then I decided to destroy the containers again and redeploy | 12:48 |
| *** DanyC has quit IRC | 12:48 | |
| Matias | then it stopped working.. | 12:48 |
| Matias | then I decided to put some debug prints in the ceilometer-collector service | 12:48 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 12:48 |
| Matias | it seems like the gnocchi dispatcher is being created, but the method which receives the samples is never called | 12:48 |
| *** Fenuks has quit IRC | 12:49 | |
| Matias | i.e. this is called: https://github.com/openstack/ceilometer/blob/master/ceilometer/dispatcher/gnocchi.py#L196 | 12:51 |
| Matias | but this is not: https://github.com/openstack/ceilometer/blob/master/ceilometer/dispatcher/gnocchi.py#L307 | 12:51 |
| *** woodard has joined #openstack-ansible | 12:52 | |
| *** woodard has quit IRC | 12:52 | |
| *** woodard has joined #openstack-ansible | 12:53 | |
| *** Jeffrey4l has quit IRC | 12:59 | |
| *** Jeffrey4l has joined #openstack-ansible | 12:59 | |
| *** rk4n has quit IRC | 13:03 | |
| *** rk4n has joined #openstack-ansible | 13:05 | |
| *** higster_ has quit IRC | 13:06 | |
| drifterza | Matias, I always had to do "ps auxwfn | grep -i ceilometer | awk '{print $2}' | xargs kill -9; service ceilometer-collector restart" to get metrics to work, this is only on new deployment. | 13:07 |
| cloudnull | mornings | 13:07 |
| drifterza | its almost as if the ceilometer-collector agent hangs on first start | 13:07 |
| drifterza | once I restarted the ceilometer-collector-agent I got new metrics etc.. | 13:08 |
| drifterza | hey cloudnull | 13:09 |
| cloudnull | o/ | 13:09 |
| *** janki has joined #openstack-ansible | 13:09 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-tests: Remove unused hostname var from test playbooks https://review.openstack.org/368829 | 13:09 |
| *** rk4n has quit IRC | 13:09 | |
| cloudnull | automagically: RE: https://review.openstack.org/#/c/354373/ -- I started in on that to see what the issues were and it seems to be a transient issue when it's not delegating to the right host and causing http://logs.openstack.org/73/354373/24/check/gate-openstack-ansible-lxc_container_create-ansible-func-ubuntu-trusty/3bde189/console.html#_2016-09-12_12_54_06_119328 | 13:11 |
| cloudnull | IDK what that is TBH | 13:11 |
| cloudnull | but it seems close to success all distros seem to be passing just not all at the same time. | 13:12 |
| automagically | cloudnull: Yeah, that’s confusing to say the least | 13:12 |
| cloudnull | trying to recreate that has been troubling too | 13:13 |
| *** openstackgerrit has quit IRC | 13:19 | |
| *** TxGirlGeek has joined #openstack-ansible | 13:20 | |
| *** openstackgerrit has joined #openstack-ansible | 13:20 | |
| *** rk4n has joined #openstack-ansible | 13:20 | |
| automagically | cloudnull: I’m sure you’ve seen this: https://github.com/ansible/ansible/issues/13773 | 13:23 |
| cloudnull | I've not | 13:25 |
| cloudnull | that may actually be it | 13:25 |
| *** neilus_ has quit IRC | 13:27 | |
| *** neilus_ has joined #openstack-ansible | 13:27 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add ansible_host to dynamic inventory https://review.openstack.org/368386 | 13:27 |
| openstackgerrit | Merged openstack/openstack-ansible-ops: implement minimal metric collection https://review.openstack.org/366167 | 13:29 |
| *** smatzek has quit IRC | 13:32 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_keystone: Remove requiretty for sudo on centos https://review.openstack.org/368850 | 13:35 |
| *** klamath_ has joined #openstack-ansible | 13:35 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 13:36 |
| cloudnull | automagically: ^ giving it a try in the gate . | 13:36 |
| cloudnull | trusty seems happy locally | 13:36 |
| *** askb has quit IRC | 13:37 | |
| evrardjp | andymccr: do you know if the swift_proxy group is defined by default in our inventories in mitaka? | 13:38 |
| evrardjp | or do the deployer needs to copy some specific env.d and therefore, there are deployments with mitaka that can have the swift_proxy group undefined | 13:38 |
| andymccr | evrardjp: ive never heard of that being an issue - we definitely define it as far as i know | 13:39 |
| evrardjp | ok | 13:39 |
| evrardjp | thanks | 13:40 |
| *** jperry has joined #openstack-ansible | 13:42 | |
| *** rk4n has quit IRC | 13:44 | |
| *** rk4n has joined #openstack-ansible | 13:45 | |
| *** Mudpuppy_afk has joined #openstack-ansible | 13:45 | |
| *** rk4n has quit IRC | 13:46 | |
| *** karimb has quit IRC | 13:49 | |
| *** jwitko has joined #openstack-ansible | 13:51 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_sahara: Convert role testing to use Ansible 2.1.1 https://review.openstack.org/353612 | 13:53 |
| *** ametts has joined #openstack-ansible | 13:54 | |
| *** smatzek has joined #openstack-ansible | 13:55 | |
| *** smatzek_ has joined #openstack-ansible | 13:56 | |
| *** michaelgugino has joined #openstack-ansible | 13:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Configure qemu when Calico network type is set https://review.openstack.org/368285 | 13:59 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 14:00 |
| *** smatzek has quit IRC | 14:00 | |
| *** BjoernT has joined #openstack-ansible | 14:00 | |
| *** Mudpuppy_afk has quit IRC | 14:01 | |
| *** jmckind has joined #openstack-ansible | 14:03 | |
| *** Mudpuppy_afk has joined #openstack-ansible | 14:03 | |
| *** Mudpuppy_afk has quit IRC | 14:04 | |
| admin0 | good read http://blog.rackspace.com/support-for-mitaka-rackspace-private-cloud-v13 | 14:04 |
| *** johnmilton has quit IRC | 14:04 | |
| cloudnull | o/ admin0 | 14:04 |
| admin0 | so what you release as stable is applied only 5 months down the line for your own private managed cloud :D ? | 14:04 |
| admin0 | so what is there today is the most stable we get ;) ? | 14:04 |
| cloudnull | not at all. | 14:05 |
| cloudnull | that's just the RPC product. | 14:05 |
| *** jmckind_ has joined #openstack-ansible | 14:05 | |
| cloudnull | OSA will continue working on stability and enhancements throughout the cycle. | 14:05 |
| admin0 | is it 5 months to upgrade ALL the enviornments combined to mitaka ? | 14:06 |
| *** jmckind_ has quit IRC | 14:07 | |
| admin0 | ignore me :D | 14:07 |
| *** jmckind has quit IRC | 14:08 | |
| *** jmckind has joined #openstack-ansible | 14:08 | |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible: Make the file name for user_secrets a variable https://review.openstack.org/368156 | 14:12 |
| *** jperry has quit IRC | 14:14 | |
| *** sc68cal_ is now known as sc68cal | 14:14 | |
| *** jperry has joined #openstack-ansible | 14:14 | |
| *** johnmilton has joined #openstack-ansible | 14:14 | |
| *** johnmilton has quit IRC | 14:14 | |
| *** johnmilton has joined #openstack-ansible | 14:15 | |
| *** jmckind_ has joined #openstack-ansible | 14:16 | |
| *** jmckind has quit IRC | 14:17 | |
| cloudnull | admin0: idk to be honest. | 14:18 |
| cloudnull | also i think this is only for greenfield deployment.s | 14:18 |
| *** BjoernT has quit IRC | 14:26 | |
| *** ddaskal has joined #openstack-ansible | 14:28 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 14:28 |
| qwang | morning, osa | 14:30 |
| *** galstrom_zzz is now known as galstrom | 14:30 | |
| cloudnull | mornings | 14:32 |
| *** michauds has joined #openstack-ansible | 14:33 | |
| openstackgerrit | Ian Cordasco proposed openstack/openstack-ansible-os_nova: Nova role should not default vars outside the nova namespace https://review.openstack.org/366881 | 14:33 |
| *** 7F1ABXRAD is now known as nishpatwa_ | 14:33 | |
| *** nishpatwa_ is now known as nish_ | 14:33 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 14:35 |
| *** jperry has quit IRC | 14:36 | |
| *** itsuugo has quit IRC | 14:37 | |
| *** jperry has joined #openstack-ansible | 14:37 | |
| *** itsuugo has joined #openstack-ansible | 14:38 | |
| *** ddaskal has quit IRC | 14:38 | |
| *** nish_ is now known as nishpatwa_ | 14:39 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 14:40 |
| *** janki has quit IRC | 14:43 | |
| *** BjoernT has joined #openstack-ansible | 14:44 | |
| openstackgerrit | Merged openstack/openstack-ansible: Compress all gathered logs for CI https://review.openstack.org/366570 | 14:50 |
| *** itsuugo has quit IRC | 15:01 | |
| *** itsuugo has joined #openstack-ansible | 15:03 | |
| *** rgogunskiy has quit IRC | 15:07 | |
| izaakk | Morning | 15:07 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Remove assumption that the neutron_lbaas var is set https://review.openstack.org/368905 | 15:07 |
| cloudnull | morning izaakk | 15:07 |
| izaakk | Hello guys, I think the docs website is down D: | 15:07 |
| izaakk | http://docs.openstack.org/developer/openstack-ansible/upgrade-guide/upgrade-playbooks.html | 15:07 |
| cloudnull | http://docs.openstack.org/developer/openstack-ansible/ | 15:09 |
| cloudnull | http://docs.openstack.org/developer/openstack-ansible/upgrade-guide/index.html | 15:09 |
| cloudnull | http://docs.openstack.org/developer/openstack-ansible/upgrade-guide/reference-upgrade-playbooks.html | 15:09 |
| *** gouthamr has joined #openstack-ansible | 15:09 | |
| cloudnull | izaakk: looks like the names changed in newton from mitaka | 15:09 |
| cloudnull | izaakk: are you at the OpenStack Hackathon in MX? | 15:10 |
| izaakk | Got it, thanks! cloudnull | 15:12 |
| cloudnull | asettle: http://docs.openstack.org/developer/openstack-ansible/upgrade-guide/reference-upgrade-playbooks.html -- that doesn't look right ? | 15:12 |
| izaakk | cloudnull: Nope, back in San Antonio, but I was there on friday and a couple hours on Saturday | 15:12 |
| cloudnull | if you have a moment can you have a look when you get a chance? | 15:12 |
| asettle | cloudnull: looking | 15:12 |
| asettle | cloudnull: anything specific you'd like to point out to me? We admittedly have not gone through the Upgrade Guide other than a cursory "oh this was mispelt" | 15:13 |
| cloudnull | asettle: it looks like the links under lbaas-version-check.yml | 15:14 |
| cloudnull | are being nested. | 15:14 |
| asettle | cloudnull: yeah okay, figured, haha. I wasn't sure if that's what you wanted me to look at. I'll look at the file and see what's up :) | 15:14 |
| cloudnull | tyvm :) | 15:14 |
| cloudnull | ohai btw | 15:15 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 15:16 |
| asettle | cloudnull: heyyy | 15:17 |
| asettle | :p | 15:17 |
| * asettle looks at the calamity that is the upgrade guide | 15:17 | |
| asettle | Hmmm | 15:17 |
| asettle | odyssey4me: next project! | 15:17 |
| asettle | The upgrade guide! | 15:17 |
| *** v1k0d3n has joined #openstack-ansible | 15:17 | |
| asettle | Good catch cloudnull - it was nested. | 15:19 |
| v1k0d3n | hey guys. have an AIO development environment that i want to upgrade from 13.1.4 > 13.3.1. i have two galera members that are in a wonky state (can't trust them) and am backing that up now. if update/upgrade apt and then use the run-upgrade.sh script...will that upgrade an AIO? | 15:19 |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: [DOCS] Reorders ToC for upgrade guide https://review.openstack.org/368916 | 15:20 |
| v1k0d3n | or do i need to rerun as if installing over again (using the aio script)? | 15:20 |
| odyssey4me | v1k0d3n no, run-upgrade is currently made for major upgrades | 15:20 |
| v1k0d3n | ok. that's what i thought, and why i asked. | 15:20 |
| odyssey4me | v1k0d3n http://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/app-minorupgrade.html | 15:21 |
| cloudnull | v1k0d3n: point release upgardes are really just a matter of checking out and reruning the parts you need/want. | 15:21 |
| cloudnull | thanks asettle | 15:21 |
| asettle | No problem cloudnull :) thanks for bringing it to my attention | 15:21 |
| odyssey4me | it would probably be a good idea for us to actually make run-upgrade work for both | 15:21 |
| v1k0d3n | ok, that makes total sense. so i can rerun the aio scripts again really for a dev env then, right? | 15:21 |
| v1k0d3n | odyssey4me: how hard would it be to make it work for both upgrade types? | 15:22 |
| odyssey4me | v1k0d3n no idea - I expect the ony bits we'd need to modify are the initial variable backups/conversion bits | 15:22 |
| odyssey4me | ideally we should detect the source tag number, then use that as the backup folder name instead of the series name | 15:23 |
| cloudnull | v1k0d3n: when I upgrade an env, i run the playbooks directly. | 15:23 |
| odyssey4me | then the rest should work just fine | 15:23 |
| cloudnull | openstack-ansible setup-everything.yml | 15:23 |
| v1k0d3n | as in a conditional declaration/variable? | 15:23 |
| cloudnull | ^ i dont generally call that playbook directly I call the individual parts. | 15:23 |
| cloudnull | but it gets the point across. | 15:24 |
| odyssey4me | v1k0d3n within a major release there aren't supposed to be changes that require restarting containers... so you can just run the playbooks in their sequence and use --limits to just control the service outages across the API's | 15:25 |
| *** admin0 has quit IRC | 15:25 | |
| v1k0d3n | makes sense. | 15:26 |
| odyssey4me | Ideally we'll be moving towards implementing that kind of rolling upgrade in Ocata in the upgrade scripts... and hopefully will switch to using playbooks instead of a mix of playbooks and scripts. | 15:26 |
| odyssey4me | but we've yet to really dig into whether that's entirely appropriate just yet | 15:27 |
| *** jheroux has joined #openstack-ansible | 15:27 | |
| *** rromans_ is now known as rromans | 15:27 | |
| *** TxGirlGeek has quit IRC | 15:29 | |
| *** pester has quit IRC | 15:34 | |
| *** Jeffrey4l has quit IRC | 15:35 | |
| *** Jeffrey4l has joined #openstack-ansible | 15:36 | |
| *** itsuugo has quit IRC | 15:42 | |
| *** itsuugo has joined #openstack-ansible | 15:44 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure file modes are 644 for inventory and group_vars for Magnum https://review.openstack.org/367164 | 15:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 15:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Correct Magnum issues found in AIO testing https://review.openstack.org/367596 | 15:48 |
| asettle | cloudnull: this is just ready and waiting for your signed approval - https://review.openstack.org/#/c/368916/ | 15:49 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_keystone: Fix depreciation "Using bare variables" https://review.openstack.org/368940 | 15:51 |
| evrardjp | I'm off for today | 15:52 |
| automagically | later evrardjp | 15:52 |
| cloudnull | thanks automagically | 15:55 |
| cloudnull | ** asettle | 15:55 |
| cloudnull | thank you too automagically :) | 15:55 |
| asettle | cloudnull: no stress | 15:55 |
| mgariepy | cloudnull, odyssey4me https://review.openstack.org/#/c/368850/ should we add this to the lxc template creation instead ? | 15:55 |
| mgariepy | mhayden, ^^ | 15:56 |
| cloudnull | I'd say yes. | 15:56 |
| * mhayden ganders | 15:56 | |
| cloudnull | that makes a lot of sense to be in the main template to effect all cent type containers. | 15:56 |
| *** pcaruana has quit IRC | 15:56 | |
| odyssey4me | mgariepy well, I think that it may reduce the overall security posture to have it changed for all containers | 15:57 |
| odyssey4me | so it's probably best to only do it where it's necessary | 15:57 |
| odyssey4me | but I'll defer to mhayden's view | 15:57 |
| mhayden | well, if i am reading this right, it would do it only on keystone containers with yum as the package mgr | 15:57 |
| mhayden | that's fairly narrow | 15:57 |
| odyssey4me | yeah, so for me this change is good | 15:57 |
| mgariepy | all os-service will probably need it | 15:58 |
| mhayden | is the root problem the fact that keystone can't sudo? | 15:58 |
| odyssey4me | well, any service doing rsync replication I expect - which means repo, nova, keystone... and that's it I think? | 15:58 |
| mgariepy | glance also need it tasks/glance_db_setup.yml | 15:58 |
| *** michaelgugino has quit IRC | 15:58 | |
| cloudnull | thats unset in the ubuntu sudoers file by default. | 15:58 |
| Matias | drifterza: already tried to kill and restart, it didn't help :( | 15:59 |
| odyssey4me | oh I see - even the sudo by ansible isn't working | 15:59 |
| cloudnull | so IMO it'd be ok to make them similar. | 15:59 |
| odyssey4me | yeah, then perhaps the lxc_hosts role should have this task as part of the cache prep | 15:59 |
| mgariepy | on ubuntu requiretty is unset and default to to not requiretty | 15:59 |
| Matias | drifterza: I was restarting it all the time, as I was trying to add debug prints to the ceilometer-collector | 15:59 |
| drifterza | It was working in mitaka | 16:00 |
| drifterza | not sure on the latest tag, haven't tested. | 16:00 |
| cloudnull | this is baffeling... https://review.openstack.org/#/c/354373/ | 16:00 |
| Matias | drifterza: however even in a fresh deployment (destroyed zookeeper_all, ceilometer_all, gnocchi_all, removed all swift containers from the gnocchi project, dropped the gnocchi database in galera, ran memcached_flush.yml) | 16:00 |
| cloudnull | if someone has time to look into it it'd be great! | 16:00 |
| Matias | it does not work to kill & restart | 16:00 |
| drifterza | is your gnocchi backend swift ? | 16:00 |
| Matias | yes | 16:00 |
| drifterza | odd. | 16:01 |
| odyssey4me | cloudnull is that where the delegated actions are sometimes delegating incorrectly? | 16:01 |
| cloudnull | yea thats what it seems like | 16:01 |
| Matias | drifterza: I was going to use ceph, but as it would need some effort in the playbooks, I choose to use swift | 16:01 |
| drifterza | I definately know it worked in mitaka dude. | 16:01 |
| Matias | which was already implemented | 16:01 |
| drifterza | ceilometer/gnocchi/swift | 16:01 |
| cloudnull | odyssey4me: do you have some magic to make that happier. | 16:01 |
| cloudnull | 's/./?/' | 16:02 |
| Matias | drifterza: mitaka had gnocchi? | 16:02 |
| Matias | I mean, in OSA | 16:02 |
| drifterza | no, I had to put that in manually. | 16:02 |
| Matias | ah | 16:02 |
| drifterza | I can give latest a stab and try troubleshoot. | 16:03 |
| mgariepy | mhayden, fatal: [aio1_glance_container-fc5998e3]: FAILED! => {"changed": false, "failed": true, "invocation": {"module_name": "command"}, "module_stderr": "sudo: sorry, you must have a tty to run sudo\n", "module_stdout": "", "msg": "MODULE FAILURE", "parsed": false} | 16:03 |
| cloudnull | odyssey4me: it doesn't look like a provider issue | 16:03 |
| Matias | I tried to do that, but as I ran into issues with upper requirements, and newton is nearly stable, I decided to try newton | 16:03 |
| mgariepy | that's the error it fixes when requiretty is needed for sudo ;) | 16:03 |
| drifterza | I might as well try Matias, I need to in any event. :) | 16:03 |
| Matias | hehehe nice | 16:03 |
| drifterza | just finished my prometheus plays with grafana, so I can give it a stab again. | 16:04 |
| Matias | grafana was nice, when I got things working | 16:04 |
| mhayden | mgariepy: that's narrow enough for me for now | 16:04 |
| Matias | I think I got it working 2 times, but I have no idea what I did of different | 16:05 |
| odyssey4me | cloudnull automagically I wonder if it relates to https://github.com/ansible/ansible-modules-extras/issues/2577 | 16:05 |
| Matias | drifterza: but anyways, did I understood the architecture correctly? the collectors are the ones which should contact gnocchi, right? | 16:05 |
| Matias | drifterza: also, they would contact via public API, not via RPC/rabbit, right? | 16:06 |
| cloudnull | odyssey4me: no. its failing to create the dirs within the container. | 16:06 |
| odyssey4me | orly? | 16:06 |
| Matias | drifterza: so I should be able to see some connections to the gnocchi port in netstat | 16:06 |
| odyssey4me | lemme look at the logs | 16:06 |
| drifterza | yeah so, you got the ceilometer agents that send to the ceilometer collector and then it sends to gnocchi | 16:06 |
| cloudnull | it goes to use the lxc module from the delegated task and cant import lxc | 16:07 |
| drifterza | so ceilometer collector to gnocchi is where the data transfers | 16:07 |
| cloudnull | wich means its executing against the wrong python, or has been delegated to the wrong location | 16:07 |
| drifterza | and thats over api not rpc | 16:07 |
| Matias | drifterza: which is odd, since the only traffic I was able to see coming out of the container is ampqs or ssh (from my ssh session) | 16:07 |
| *** jperry has quit IRC | 16:07 | |
| *** jperry has joined #openstack-ansible | 16:08 | |
| Matias | drifterza: but maybe it is not collecting from the rabbit queue in the first place, which would explain why it didn't make contact with gnocchi at all | 16:08 |
| odyssey4me | cloudnull it appears to be using the right python: http://logs.openstack.org/73/354373/30/check/gate-openstack-ansible-lxc_container_create-ansible-func-ubuntu-trusty/97ad300/console.html#_2016-09-12_15_55_44_032683 | 16:09 |
| drifterza | Matias, thats is most likely whats happening | 16:09 |
| Matias | drifterza: I was going to study the ceilometer code to try to find where it collects the data from the agents, but if you have some pointer I would be very grateful | 16:09 |
| mgariepy | mhayden, it will probably affect all the tasks that have become: yes in them. | 16:11 |
| drifterza | I assume you have the ceilometer_gnocchi_enabled: true etc.. | 16:11 |
| drifterza | and stuff like : nova_ceilometer_enabled: true | 16:12 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 16:13 |
| cloudnull | odyssey4me: i recently set that option to pin it to /usr/bin/python2 -- i just unset that | 16:13 |
| Matias | drifterza: yes | 16:13 |
| Matias | drifterza: in two deploys I did in the same hw, with the same config and the same git rev, it did work | 16:14 |
| odyssey4me | cloudnull I wonder if https://review.openstack.org/#/c/354373/30/tests/inventory is resulting in 'localhost' being misinterpreted somehow... perhaps ansible_connecton=local is necessary to ensure it's done on the host? | 16:14 |
| Matias | the problem is it is not very reproducible, it never works in the first deploy, I have to randomly destroy and recreate containers until it works | 16:14 |
| drifterza | I forget in the plays do we SCM the master commit or a ref commit ? | 16:14 |
| odyssey4me | otherwise it might be trying to connect to localhost, which could be any of the containers | 16:14 |
| Matias | drifterza: master commit of the roles when you run bootstrap-ansible, and always a ref commit of nova,neutron,ceilo,etc. code | 16:15 |
| cloudnull | odyssey4me: i did try that too, no love. | 16:15 |
| drifterza | yeah could be something weird in the master for gnocchi dude. | 16:16 |
| cloudnull | the playbooks have the connection plugin they're supposed to use. | 16:16 |
| drifterza | I would have to deploy and check | 16:16 |
| cloudnull | and pinning it had no effect. | 16:16 |
| cloudnull | its almost like its not executing with sude | 16:16 |
| cloudnull | *sudo | 16:16 |
| drifterza | anyways, I have to go home. Its late here. GL Matias | 16:17 |
| Matias | drifterza: cya, thanks | 16:17 |
| Matias | drifterza: 01:17 there? | 16:18 |
| Matias | drifterza: have a good sleep :P | 16:18 |
| Matias | no, it's worse.. 04:18 | 16:19 |
| *** drifterza has quit IRC | 16:21 | |
| Matias | cloudnull: regarding https://bugs.launchpad.net/openstack-ansible/+bug/1620765 I think the repo_build role is not the best place to put the call to the file module to recursively change the git repos owner, because it runs only in one of the repo containers | 16:22 |
| openstack | Launchpad bug 1620765 in openstack-ansible "Multiple runs of repo-build break git repo replication" [Undecided,In progress] - Assigned to Paulo Matias (paulo-matias) | 16:22 |
| *** spotz_zzz is now known as spotz | 16:22 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 16:23 |
| *** thorst has quit IRC | 16:23 | |
| Matias | cloudnull: would it be bad practice to put it in the repo-build.yml playbook in order to get it to run in all containers? | 16:23 |
| cloudnull | odyssey4me: i added a few `become: true` it looks like with delegation its not executing sudo | 16:23 |
| cloudnull | and in the gate its running as the jenkins user | 16:23 |
| *** KLevenstein has joined #openstack-ansible | 16:24 | |
| cloudnull | Matias: adding it as a post_task makes sense to me | 16:24 |
| odyssey4me | cloudnull it definitely looks like it's the tox python versus the host python | 16:24 |
| odyssey4me | http://logs.openstack.org/73/354373/31/check/gate-openstack-ansible-lxc_container_create-ansible-func-ubuntu-xenial/2a0f322/console.html#_2016-09-12_16_21_02_421810 | 16:24 |
| odyssey4me | versus the working one: http://logs.openstack.org/73/354373/31/check/gate-openstack-ansible-lxc_container_create-ansible-func-ubuntu-trusty/e6b17ca/console.html#_2016-09-12_16_20_14_821632 | 16:24 |
| cloudnull | yea in that case we see "The `lxc` module is not importable. Check the requirements." | 16:25 |
| cloudnull | where in the other its not sudo'ing | 16:25 |
| cloudnull | so maybe we need both. :) | 16:25 |
| Matias | cloudnull: I don't know if post_task would work, suppose you have repo-01, repo-02, repo-03; then in the first time it ran repo_build inside repo-01, files there would be owned by root; in the second time, if it runs in repo-02, repo-01 would start to try synchronizing these files with lsyncd and wouldn't succeed | 16:26 |
| odyssey4me | well, on xenial (where it's not working) it's looking for the module in the tox venv, and it won't find it there | 16:26 |
| odyssey4me | we had this issue before | 16:26 |
| Matias | cloudnull: looking by this angle, maybe pre_task makes more sense | 16:26 |
| cloudnull | odyssey4me: its not working on trusty either. | 16:26 |
| cloudnull | but sometimes it works on xenial and trusty | 16:27 |
| Matias | unless you wait for lsyncd to retry, which I think it does... then it does not make any difference | 16:27 |
| odyssey4me | cloudnull but the trust/centos jobs had a successful result in that run | 16:27 |
| cloudnull | then we ran recheck and all three failed | 16:27 |
| odyssey4me | I know it's not working *consistently* | 16:27 |
| odyssey4me | last time this was an issue it related to the sudoers config on the infra image | 16:28 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 16:28 |
| v1k0d3n | odyssey4me cloudnull running into an issue with the repo container again for some reason. | 16:30 |
| *** sdake_ has joined #openstack-ansible | 16:30 | |
| v1k0d3n | this happened to me before...not sure exactly what was causing it. "/opt/op-clone-script.sh" fails | 16:31 |
| Matias | v1k0d3n: could it be because of this? https://bugs.launchpad.net/openstack-ansible/+bug/1620765 | 16:31 |
| openstack | Launchpad bug 1620765 in openstack-ansible "Multiple runs of repo-build break git repo replication" [Undecided,In progress] - Assigned to Paulo Matias (paulo-matias) | 16:31 |
| Matias | v1k0d3n: please try to change the owner manually to see if it fixes | 16:32 |
| cloudnull | v1k0d3n: ++ what Matias said. | 16:32 |
| cloudnull | odyssey4me: do you think there's different configs within the sudoers file on the infra images? | 16:33 |
| odyssey4me | cloudnull and yes, I see there's a lack of 'echo BECOME-SUCCESS-hrloskzgehgspzokrglvsghzgwurljxr;' in the EXEC line | 16:33 |
| odyssey4me | so it does appear that it is not executing via sudo, which appears to be the root cause here | 16:33 |
| v1k0d3n | ah...yes, had this issue on another deployment also and it drove me nuts. | 16:33 |
| cloudnull | its possible that ansible folks broke delegate_to and become once again ? | 16:33 |
| v1k0d3n | remember cloudnull i was talking jibberish about repo issues about a month or so ago? | 16:34 |
| v1k0d3n | lol | 16:34 |
| odyssey4me | cloudnull if it was ansible, surely it'd be consistent? | 16:34 |
| Matias | ansible, the only single piece of software which has more bugs than openstack itself | 16:34 |
| Matias | :P | 16:34 |
| cloudnull | there's so much jibberish in my memory :) | 16:34 |
| cloudnull | its hard to keep track | 16:35 |
| cloudnull | you'd think | 16:35 |
| v1k0d3n | hahah! fair enough. that seems to be a problem everywhere. :) | 16:35 |
| cloudnull | could be a race | 16:35 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] [Docs] Metadata cleanup https://review.openstack.org/368957 | 16:35 |
| v1k0d3n | Matias: change user to what? | 16:35 |
| odyssey4me | cloudnull yeah, that'd be my guess - especially if we're seeing inconsistent results from the same provider | 16:35 |
| *** hughmFLEXin has joined #openstack-ansible | 16:36 | |
| v1k0d3n | Matias: oh nginx? | 16:37 |
| *** thorst_ has joined #openstack-ansible | 16:37 | |
| *** jmckind has joined #openstack-ansible | 16:37 | |
| *** michaelgugino has joined #openstack-ansible | 16:38 | |
| *** jmckind_ has quit IRC | 16:38 | |
| Matias | v1k0d3n: yep | 16:39 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 16:40 |
| *** hughmFLEXin has quit IRC | 16:40 | |
| cloudnull | odyssey4me: so it got farther with adding ``become: true`` so i added that to all of the tasks with a delegation . | 16:40 |
| cloudnull | we'll see | 16:40 |
| *** hughmFLEXin has joined #openstack-ansible | 16:40 | |
| *** jmckind has quit IRC | 16:42 | |
| hughmFLEXin | hey guys im running into this error while running setup-hosts.yml | 16:43 |
| hughmFLEXin | nlxc-info: confile.c: config_string_item_max: 341 f3e05dd2_eth2.41 is too long (>= 16)\nlxc-info: parse.c: lxc_file_for_each_line: 57 Failed to parse config: lxc.network.veth.pair = f3e05dd2_eth2.41\n\nlxc-info: parse.c: lxc_file_for_each_line: 57 Failed to parse config: lxc.include = /var/lib/lxc/infra2_neutron_server_container-f3e05dd2/eth2.41.ini\n\nFailure to retrieve information on /var/lib/lxc:infra2_neutron_server_co | 16:43 |
| hughmFLEXin | ntainer-f3e05dd2", "stdout": "", "stdout_lines": [], "warnings": []} | 16:43 |
| v1k0d3n | Matias: am i changing the www dir to nginx:www-data recursively manually? | 16:43 |
| v1k0d3n | sorry for the ignorance. tried ^^ and failed the same. | 16:43 |
| cloudnull | hughmFLEXin: vethpair name it too long. | 16:44 |
| *** spotz is now known as spotz_zzz | 16:45 | |
| cloudnull | did you intend to give the container tag interface? | 16:45 |
| *** ametts has quit IRC | 16:45 | |
| hughmFLEXin | cloudnull: yes | 16:45 |
| cloudnull | hum... i dont think we've ever encountered this issue. | 16:45 |
| hughmFLEXin | cloudnull: does the container interface have to correlate exactly as the actual interface? or does it not matter | 16:46 |
| cloudnull | it doesn't matter. but in our role we take the exact interface given to the container, use the conatiner hash + interface name as the named veth. | 16:46 |
| cloudnull | which is used to identify what is connected | 16:47 |
| cloudnull | good troubshooting management, metrics, etc | 16:47 |
| cloudnull | but we should fense around names being too long and it seems we've missed a case. | 16:47 |
| cloudnull | which is done here https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/tasks/container_create.yml#L253 | 16:48 |
| hughmFLEXin | yea, i'm working off a sample interface file from @jamesdenton, | 16:48 |
| cloudnull | and here https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/templates/container-interface.ini.j2#L11 | 16:49 |
| odyssey4me | IIRC the interface name must be no more than ethXX (5 chars) | 16:49 |
| v1k0d3n | Matias: ah, looked at the change and manually updated the repo_clone_git.yml. trying again. thanks for the heads up. | 16:49 |
| cloudnull | hughmFLEXin: i just double checked and the "length" a veth can have is 15 | 16:50 |
| cloudnull | in your case its 16 | 16:50 |
| odyssey4me | cloudnull looks like the handlers will need it too? http://logs.openstack.org/73/354373/34/check/gate-openstack-ansible-lxc_container_create-ansible-func-ubuntu-xenial/be77507/console.html#_2016-09-12_16_47_15_378946 | 16:51 |
| cloudnull | :\ | 16:51 |
| odyssey4me | cloudnull surely the playbook execution should be done with become: true and become_user: root | 16:52 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 16:52 |
| cloudnull | it shouldn't have to because those are set in inventory | 16:53 |
| cloudnull | and "should" override the bits in the playbook . | 16:53 |
| cloudnull | which is frustrating because sometimes it is | 16:53 |
| odyssey4me | yep, but in the inventory here we have no bceom_user set | 16:53 |
| odyssey4me | for localhost | 16:53 |
| hughmFLEXin | is that a lxc limitation or is that a osa imposed limitation | 16:54 |
| cloudnull | hughmFLEXin: its a linux kernel limitation | 16:54 |
| hughmFLEXin | i guess a workaround would be for me to shorten the inventory hostname being concatenated? | 16:55 |
| hughmFLEXin | in the playobok | 16:55 |
| cloudnull | yea you can shorten it, which is what I think im going to do as a set fact. | 16:56 |
| cloudnull | that way it sets a var and limits the var size | 16:56 |
| odyssey4me | cloudnull would this not be resolved by setting become_user in https://review.openstack.org/#/c/354373/35/tests/inventory for localhost? | 16:57 |
| hughmFLEXin | which brings up the next q.....is there any clean up that needs to be done due to the failure, or can i just rerun | 16:57 |
| cloudnull | if you make the change manually you can just rerun | 16:57 |
| cloudnull | odyssey4me: i dont think so | 16:58 |
| cloudnull | https://github.com/ansible/ansible/blob/devel/lib/ansible/constants.py#L225 | 16:58 |
| cloudnull | become_user=root is already the default | 16:58 |
| hughmFLEXin | perfect...will make changes locally and give it a shot, thanks cloudnull, ill update you later if your around | 16:58 |
| *** neilus_ has left #openstack-ansible | 16:58 | |
| cloudnull | hughmFLEXin: i'll be around. | 16:58 |
| cloudnull | odyssey4me: so far trusty and xenial are both success. | 16:59 |
| *** TxGirlGeek has joined #openstack-ansible | 17:00 | |
| odyssey4me | cloudnull ok, so perhaps winding it back and trying it with the become set in the playbook will do it | 17:00 |
| odyssey4me | clearly the inventory application isn't working | 17:00 |
| cloudnull | yea. sadly. | 17:01 |
| cloudnull | i guess we can try setting it here https://github.com/openstack/openstack-ansible-tests/blob/master/test-prepare-containers.yml | 17:01 |
| odyssey4me | yep, and we should be removing the container_release and backing_store vars too | 17:02 |
| odyssey4me | I thin kit should also be removing the global_env_vars - that was only set for testing as I recall | 17:03 |
| hughmFLEXin | cloudnull: looks like i have to fork the container_create role and change out the git path in role_requirements.txt? | 17:05 |
| cloudnull | odyssey4me: that was from the tests repo | 17:06 |
| *** sdake has joined #openstack-ansible | 17:06 | |
| odyssey4me | cloudnull yep, I know - I'm saying that all tests should not be using those vars | 17:07 |
| odyssey4me | only the lxc-container-create role tests should be using those vars | 17:07 |
| cloudnull | and thats a copy of https://review.openstack.org/#/c/354373/35/tests/test-prepare-containers.yml | 17:07 |
| cloudnull | so i guess an update there is in order. | 17:07 |
| cloudnull | which is neat. because that means we've been testing mixed container envs | 17:08 |
| cloudnull | :) | 17:08 |
| cloudnull | hughmFLEXin: kinda. | 17:08 |
| cloudnull | you can fork it | 17:08 |
| cloudnull | or modify it | 17:08 |
| cloudnull | the role is stored at /etc/ansible/roles | 17:08 |
| cloudnull | if you want to get out of this jam, I'd just modify it locally and rerun the playbook | 17:08 |
| hughmFLEXin | ah ok, i never used roles in that manner, so i was wondering if the roles were automagically being stored in memory ors omething | 17:09 |
| *** sdake_ has quit IRC | 17:09 | |
| *** jperry has quit IRC | 17:10 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests: Remove pinned vars and set become:true https://review.openstack.org/368983 | 17:10 |
| cloudnull | odyssey4me: ^ | 17:10 |
| stevelle | Matias: there is definitely something wrong about the Ceilo service startup that needs to be fixed but we don't seem to know what exactly the fix is yet and I haven't always seen that issue but it's a long-running concern. | 17:11 |
| odyssey4me | cloudnull one more tweak | 17:12 |
| stevelle | it only makes matters worse that it is hard to inspect ceilo operation. | 17:12 |
| *** jperry has joined #openstack-ansible | 17:13 | |
| odyssey4me | cloudnull lemme tweak it and write up a commit message to explain the removal | 17:13 |
| cloudnull | tweak away my good man | 17:14 |
| *** thorst_ has quit IRC | 17:14 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Remove pinned vars and set become:true https://review.openstack.org/368983 | 17:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Remove pinned vars and set become:true https://review.openstack.org/368983 | 17:19 |
| *** avlis has quit IRC | 17:19 | |
| odyssey4me | ok, I think that's it | 17:20 |
| cloudnull | so become_allthethings https://review.openstack.org/#/c/354373/ -- success | 17:21 |
| cloudnull | hopfully setting it in the playbook makes life better. | 17:21 |
| odyssey4me | yep | 17:21 |
| cloudnull | which begs the question, do we need that here too https://github.com/openstack/openstack-ansible/blob/master/playbooks/lxc-containers-create.yml | 17:22 |
| cloudnull | or does `user:root` have some magical property which is insulating us from that issue ? | 17:23 |
| cloudnull | stevelle: is that a long running osa issue or ceilo issue? | 17:24 |
| *** itsuugo has quit IRC | 17:24 | |
| odyssey4me | cloudnull so I'm guessing that user:root is setting the use that we login to the host as - so sudo isn't necessary because it's already root | 17:24 |
| odyssey4me | if we want to switch to allowing non-root remote login, then we'll have to set 'become' on the appropriate tasks/plays | 17:25 |
| *** itsuugo has joined #openstack-ansible | 17:25 | |
| cloudnull | but jenkins in the gate is not executing at root | 17:25 |
| odyssey4me | I think it does make sense for us to move away from a fixed user for login, and instead allow any user and we rather sudo | 17:25 |
| cloudnull | are we running sudo from tox maybe? | 17:25 |
| odyssey4me | cloudnull the integrated gate runs sudo - the roles do not | 17:26 |
| *** kvcobb has joined #openstack-ansible | 17:26 | |
| cloudnull | ah . | 17:26 |
| cloudnull | ok | 17:26 |
| cloudnull | makes sense now | 17:26 |
| hughmFLEXin | cloudnull: hrm, after making the change (-8 -> -4) on the slicing, it failed before i even reached the rewiring part. | 17:26 |
| stevelle | cloudnull: osa issue imo | 17:26 |
| odyssey4me | the integrated gate is executed using: sudo scripts/gate-check-commit.sh | 17:26 |
| stevelle | until we have a detailed understanding we won't really know though cloudnull | 17:26 |
| cloudnull | stevelle: something maybe we can focus on for the RCs? | 17:26 |
| stevelle | cloudnull: it would be helpful to focus eyeballs on | 17:27 |
| hughmFLEXin | cloudnull: https://gist.github.com/bbyhuy/ad1f4ee6fd8ab06dbc41cf37198d4f57 | 17:27 |
| *** electrofelix has quit IRC | 17:27 | |
| odyssey4me | the roles are executed using tox as the jenkins user, which has sudo capabilities - but we have to execute sudo | 17:27 |
| odyssey4me | so yes - it makes sense for the test playbooks to set 'become' | 17:27 |
| cloudnull | hughmFLEXin: run: openstack-ansible lxc-container-destroy.yml and then rerun the create play | 17:28 |
| cloudnull | indeed. | 17:28 |
| *** kvcobb has quit IRC | 17:28 | |
| hughmFLEXin | cloudnull: hrm, same error running destroy | 17:29 |
| cloudnull | interesting , you mind sharing the patch ? | 17:30 |
| *** admin0 has joined #openstack-ansible | 17:30 | |
| *** kvcobb has joined #openstack-ansible | 17:32 | |
| *** jcannava has joined #openstack-ansible | 17:32 | |
| *** asettle has quit IRC | 17:33 | |
| hughmFLEXin | cloudnull: https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/tasks/container_create.yml#L253 | 17:33 |
| hughmFLEXin | same link you gave me | 17:33 |
| hughmFLEXin | only change i did was shorten the slice from 8 to 4 | 17:33 |
| *** kvcobb has quit IRC | 17:34 | |
| *** kvcobb has joined #openstack-ansible | 17:34 | |
| *** kvcobb has quit IRC | 17:38 | |
| openstackgerrit | Merged openstack/openstack-ansible: [DOCS] Reorders ToC for upgrade guide https://review.openstack.org/368916 | 17:40 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ensure veths are only 15 characters long https://review.openstack.org/368989 | 17:40 |
| cloudnull | hughmFLEXin: looking now | 17:41 |
| hughmFLEXin | np | 17:42 |
| cloudnull | interesting, if you puth the old bits back and run destroy is that functional? | 17:42 |
| hughmFLEXin | nope...i just tried that haha... | 17:42 |
| cloudnull | ha | 17:43 |
| cloudnull | great minds | 17:43 |
| cloudnull | you can do a simple lxc-destory -fn $NAME | 17:43 |
| hughmFLEXin | alright | 17:44 |
| mgariepy | cloudnull, do you mind if I patch your patch ? https://review.openstack.org/#/c/320542/ ? | 17:45 |
| cloudnull | also, if you dont mind pulling in a patch https://review.openstack.org/#/c/368989/ | 17:45 |
| cloudnull | mgariepy: i dont mind at all. | 17:45 |
| cloudnull | mgariepy: it'd be great to get some help on cent things , so have at it. | 17:45 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Add network conf auditing on CentOS https://review.openstack.org/368991 | 17:45 |
| cloudnull | hughmFLEXin: https://review.openstack.org/#/c/368989/ -- should correct the veth legth issue | 17:46 |
| v1k0d3n | Matias: hey, after a couple of attempts (and rebuilding the container) it worked. thank you for the update, and thanks cloudnull for pointing me in that direction...that was it! :) | 17:46 |
| mgariepy | cloudnull, i | 17:47 |
| *** thorst_ has joined #openstack-ansible | 17:47 | |
| cloudnull | v1k0d3n: was it the perms ? | 17:47 |
| mgariepy | cloudnull, i'm getting there, keystone is patched up, next is glance then the next playbook in setup-openstack ;) hehe | 17:47 |
| cloudnull | nice! | 17:47 |
| v1k0d3n | yeah, when i added the two additional become lines to the file mentioned in the issue...it worked. | 17:47 |
| v1k0d3n | it took me a couple of tries to figure out what needed destroyed and order of playbooks, but all good. | 17:48 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ensure veths are only 15 characters long https://review.openstack.org/368989 | 17:48 |
| cloudnull | Matias: ^ | 17:48 |
| cloudnull | ^ what v1k0d3n said | 17:48 |
| *** thorst_ has quit IRC | 17:49 | |
| cloudnull | v1k0d3n: nice its geting worked out . | 17:49 |
| cloudnull | we should get those fixes in and backported to mitaka | 17:49 |
| cloudnull | Matias odyssey4me ^ | 17:49 |
| *** thorst_ has joined #openstack-ansible | 17:49 | |
| odyssey4me | cloudnull yep, agreed | 17:51 |
| v1k0d3n | i noticed that jenkins is failing on trusty? | 17:51 |
| v1k0d3n | i'm on trusty...seems to be taking a while at another step (but waiting it out to see) | 17:51 |
| odyssey4me | v1k0d3n which test is failing on trusty? | 17:52 |
| hughmFLEXin | @cloudnull fancy patch :-P | 17:52 |
| v1k0d3n | am i reading this correctly? | 17:52 |
| v1k0d3n | https://review.openstack.org/#/c/366280/ | 17:52 |
| v1k0d3n | gate-openstack-ansible-repo_build-ansible-func-ubuntu-trusty | 17:52 |
| v1k0d3n | still truckin' on my end though...meh. | 17:53 |
| odyssey4me | yeah, that patch is failing at this point - I haven't looked into why | 17:53 |
| odyssey4me | it's failing all functional tests on all platforms | 17:53 |
| cloudnull | v1k0d3n: how goes the POC? | 17:53 |
| cloudnull | or are you beyond that now ? | 17:53 |
| v1k0d3n | well, i ran into issues....this issue in fact. | 17:57 |
| v1k0d3n | and had trouble figuring it out at the time. needed it quickly for a demo/meetup i was scheduled to do. | 17:58 |
| v1k0d3n | now kind of excited that i may be able to try again :) | 17:58 |
| *** thorst_ has quit IRC | 17:58 | |
| *** KLevenstein has quit IRC | 18:00 | |
| *** poopcat has joined #openstack-ansible | 18:01 | |
| *** berendt has quit IRC | 18:04 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Use pip_install_options when creating reqt wheels https://review.openstack.org/368172 | 18:04 |
| *** TxGirlGeek has quit IRC | 18:04 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_server: Set permissions on the openstackgit directory https://review.openstack.org/368997 | 18:10 |
| cloudnull | Matias: v1k0d3n: ^ idk if that totally solves the git perms issues | 18:10 |
| cloudnull | but its a start | 18:10 |
| *** itsuugo has quit IRC | 18:16 | |
| *** TxGirlGeek has joined #openstack-ansible | 18:17 | |
| *** itsuugo has joined #openstack-ansible | 18:17 | |
| *** catlook has joined #openstack-ansible | 18:18 | |
| catlook | howdy...i'm preparing to run through the openstack ansible install guide. in preparation, i ran through some steps to set up my test environment. The last step was running an ansible playbook. My instructions provided an inventory.ini file, and a bootstrap-hosts.yml file. WHen i run the playbook command however, the system responds: ./bootstrap-hosts.yml could not be found | 18:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Fix depreciation "Using bare variables" https://review.openstack.org/368940 | 18:21 |
| catlook | Any idea what I'm doing wrong? | 18:21 |
| *** jheroux has quit IRC | 18:22 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-ceph_client: fix basic support to pass centos install https://review.openstack.org/369004 | 18:23 |
| *** itsuugo has quit IRC | 18:23 | |
| cloudnull | catlook: what command did you run ? | 18:23 |
| cloudnull | and are you running an all in one? | 18:23 |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Remove pinned vars and set become:true https://review.openstack.org/368983 | 18:23 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 18:23 |
| *** itsuugo has joined #openstack-ansible | 18:25 | |
| catlook | the command was: ansible-playbook -i ./inventory.ini ./bootstrap-hosts.yml | 18:25 |
| catlook | both files have permissions of 644 | 18:25 |
| catlook | not sure about "all in one" question | 18:26 |
| *** kvcobb has joined #openstack-ansible | 18:28 | |
| openstackgerrit | Ian Cordasco proposed openstack/openstack-ansible-os_nova: Nova role should not default vars outside the nova namespace https://review.openstack.org/366881 | 18:28 |
| *** schwicht has joined #openstack-ansible | 18:31 | |
| qwang | catlook: is bootstrap-hosts.yml file under your current directory? | 18:35 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_rally: Ansible 2.1.x role testing https://review.openstack.org/353610 | 18:36 |
| catlook | yes | 18:37 |
| *** itsuugo has quit IRC | 18:37 | |
| catlook | both yml ans ini files are in the current dir | 18:37 |
| *** itsuugo has joined #openstack-ansible | 18:38 | |
| automagically | cloudnull Woot! https://review.openstack.org/#/c/354373/36 Nice collaboration between you and odyssey4me | 18:39 |
| automagically | To track down the root cause of the issues | 18:39 |
| *** KLevenstein has joined #openstack-ansible | 18:40 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_rally: Remove apt-get update from Vagrantfile https://review.openstack.org/369008 | 18:41 |
| *** jmckind has joined #openstack-ansible | 18:41 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_glance: Implement CentOS 7 support in os_glance https://review.openstack.org/320542 | 18:42 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_sahara: Convert role testing to use Ansible 2.1.1 https://review.openstack.org/353612 | 18:46 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_rally: Ansible 2.1.x role testing https://review.openstack.org/353610 | 18:54 |
| hughmFLEXin | cloudnull: hrm, even lxc-destroy -fn doesn't work | 18:55 |
| hughmFLEXin | cloudnull: https://gist.github.com/bbyhuy/70749acaafff5df3398974d7589aa1fa | 18:56 |
| *** thorst_ has joined #openstack-ansible | 18:56 | |
| *** itsuugo has quit IRC | 18:56 | |
| *** itsuugo has joined #openstack-ansible | 18:57 | |
| *** TxGirlGe_ has joined #openstack-ansible | 18:59 | |
| *** TxGirlGe_ has quit IRC | 19:00 | |
| *** spedione|AWAY is now known as spedione | 19:00 | |
| *** TxGirlGeek has quit IRC | 19:00 | |
| *** thorst_ has quit IRC | 19:03 | |
| *** pcaruana has joined #openstack-ansible | 19:04 | |
| *** itsuugo has quit IRC | 19:04 | |
| *** itsuugo has joined #openstack-ansible | 19:06 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [Docs] Metadata cleanup https://review.openstack.org/368957 | 19:07 |
| *** asettle has joined #openstack-ansible | 19:14 | |
| *** klamath has quit IRC | 19:16 | |
| *** klamath has joined #openstack-ansible | 19:17 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 19:20 |
| Matias | cloudnull: it looks good, I will test it | 19:21 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [Docs] Update dev guide for metadata docs https://review.openstack.org/369015 | 19:23 |
| Matias | cloudnull: my path on repo_build would still be needed though, to avoid new files owned by root to be created by the script | 19:23 |
| Matias | my patch* | 19:23 |
| Matias | cloudnull: I had set your PR to the same topic | 19:24 |
| *** spedione is now known as spedione|AWAY | 19:26 | |
| *** asettle has quit IRC | 19:26 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [Docs] Metadata cleanup https://review.openstack.org/368957 | 19:28 |
| *** asettle has joined #openstack-ansible | 19:28 | |
| *** asettle has quit IRC | 19:28 | |
| cloudnull | sorry was afk while eating. | 19:28 |
| *** asettle has joined #openstack-ansible | 19:28 | |
| cloudnull | hughmFLEXin: interesting that the lxc-destory is still a no go. | 19:28 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-galera_server: Ansible 2.1.1 role testing https://review.openstack.org/354411 | 19:29 |
| cloudnull | hughmFLEXin: id you remove "lxc.include = /var/lib/lxc/infra1_utility_container-df779a4f/eth2.41.ini" from the container config file "/var/lib/lxc/infra1_utility_container-df779a4f/config" does it still not work ? | 19:29 |
| cloudnull | Matias: yea it looks like both will still be needed. | 19:29 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Don't merge -- testing gate https://review.openstack.org/369019 | 19:29 |
| cloudnull | automagically: https://review.openstack.org/#/c/354373 -- other cores it'd be great to get this in soon-ish as it'll be a building block for all other tests. | 19:31 |
| *** asettle has quit IRC | 19:33 | |
| *** asettle has joined #openstack-ansible | 19:33 | |
| cloudnull | also https://review.openstack.org/#/c/368989/ | 19:34 |
| cloudnull | which should fix the issues reported by hughmFLEXin this morning | 19:34 |
| cloudnull | ^ that should go back to mitka imo. | 19:35 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-galera_server: Ansible 2.1.1 role testing https://review.openstack.org/354411 | 19:35 |
| *** asettle has quit IRC | 19:38 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Ansible 2.1.1 role testing https://review.openstack.org/354373 | 19:41 |
| -cloudnull- This is a rolling etherpad of items that need to be worked on -- https://etherpad.openstack.org/p/osa-newton-rc1 -- if you've worked on things please add them to the etherpda, if you've fixed things please strick through the items on the etherpad. | 19:43 | |
| -cloudnull- This is a rolling etherpad of items that need to be worked on -- https://etherpad.openstack.org/p/osa-newton-rc1 -- if you've worked on things please add them to the etherpad, if you've fixed things please strike through the items on the etherpad. | 19:43 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-galera_server: Ansible 2.1.1 role testing https://review.openstack.org/354411 | 19:44 |
| Matias | cloudnull: -rwxr-sr-x 1 nginx www-data 17 Sep 6 17:41 /var/www/repo/openstackgit/aodh/babel.cfg | 19:46 |
| Matias | files are being set +s | 19:46 |
| cloudnull | thats good :) | 19:46 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_barbican: Ansible 2.1.1 role testing https://review.openstack.org/353567 | 19:46 |
| cloudnull | so even if the user perms change they should still be accessible on a later run | 19:47 |
| cloudnull | though maybe the perms should really be 2775 | 19:47 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [Docs] Update configuration/controls docs https://review.openstack.org/369025 | 19:47 |
| openstackgerrit | Merged openstack/openstack-ansible: Correct Magnum issues found in AIO testing https://review.openstack.org/367596 | 19:47 |
| openstackgerrit | Merged openstack/openstack-ansible: Ensure file modes are 644 for inventory and group_vars for Magnum https://review.openstack.org/367164 | 19:47 |
| openstackgerrit | Merged openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 19:47 |
| Matias | semantics of g+s on files is not so that if you run some executable it runs with setgid to www-data ? | 19:47 |
| openstackgerrit | Merged openstack/openstack-ansible: Update all SHAs for 12.2.4 https://review.openstack.org/367549 | 19:47 |
| Matias | we should have g+s only on dirs | 19:48 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_sahara: Convert role testing to use Ansible 2.1.1 https://review.openstack.org/353612 | 19:48 |
| Matias | openstackgerrit merging changes like there is not tomorrow | 19:48 |
| cloudnull | Matias: is that after pulling in my patch ? | 19:48 |
| cloudnull | g+s should only be on dirs | 19:48 |
| Matias | cloudnull: yep | 19:48 |
| cloudnull | I added recusrse to the task | 19:49 |
| Matias | ansible sets on everything | 19:49 |
| cloudnull | which only effects dirs unless its ansible and doesn't really do what the docs say | 19:49 |
| Matias | I don't know if the file module has this feature | 19:49 |
| cloudnull | :p | 19:49 |
| Matias | ah | 19:49 |
| Matias | once again | 19:49 |
| Matias | recursively set the specified file attributes (applies only to state=directory) | 19:49 |
| cloudnull | ^ that | 19:50 |
| Matias | cloudnull: applies on only to *state=*directory | 19:50 |
| Matias | and not "applies only to directories" :P | 19:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally: Ansible 2.1.x role testing https://review.openstack.org/353610 | 19:50 |
| cloudnull | that should read (applies only to state=directory, unless there are files in the directory then it applis to everything) | 19:51 |
| Matias | cloudnull: http://stackoverflow.com/a/29793833 | 19:51 |
| Matias | hmm mergh that would not solve it | 19:52 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Add network conf auditing on CentOS https://review.openstack.org/368991 | 19:52 |
| Matias | "or files with at least one x bit set" | 19:52 |
| cloudnull | In my mind that should basically do (find $PATH -type d -exec chmod $MODE {} \; ) | 19:52 |
| kylek3h | odyssey4me, cloudnull, thanks for nudging this along! https://review.openstack.org/#/c/367596/ | 19:52 |
| Matias | that would set g+s on executable files, which we definetelly don't want | 19:52 |
| *** TxGirlGeek has joined #openstack-ansible | 19:52 | |
| Matias | yes, probably we need to use command | 19:52 |
| cloudnull | I think so. | 19:53 |
| Matias | the file module doesn't have this feature :/ | 19:53 |
| cloudnull | :| | 19:53 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_server: Set permissions on the openstackgit directory https://review.openstack.org/368997 | 19:54 |
| cloudnull | Matias: ^ I removed the recurse | 19:54 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_zaqar: Ansible 2.1.1 role testing https://review.openstack.org/353617 | 19:54 |
| cloudnull | kylek3h: ofcourse , thanks for pushing the PR | 19:55 |
| Matias | cloudnull: we could do two steps | 19:55 |
| Matias | cloudnull: mode without recurse | 19:55 |
| Matias | then owner with recurse | 19:55 |
| cloudnull | I think we could do it with one shell task | 19:56 |
| cloudnull | find $PATH -type d -exec chmod 02775 {} \; | 19:56 |
| Matias | the problem with shell tasks is to detect changes | 19:56 |
| cloudnull | find $PATH -type f -exec chmod 01644 {} \; | 19:56 |
| cloudnull | it's hard to detect changes there. | 19:57 |
| cloudnull | sadly. | 19:57 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-galera_server: Ansible 2.1.1 role testing https://review.openstack.org/354411 | 19:57 |
| Matias | the two-step file module usage would solve all problems, I think | 19:57 |
| Matias | g+s would already be propagated even if the user originally run it as root | 19:58 |
| Matias | unless the directory was manually created at some point, and repo-build ran on it | 19:58 |
| *** thorst_ has joined #openstack-ansible | 20:00 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Ensure veths are only 15 characters long https://review.openstack.org/368989 | 20:01 |
| openstackgerrit | Merged openstack/openstack-ansible: Properly namespace cinder_storage_address https://review.openstack.org/368653 | 20:02 |
| *** TxGirlGeek has quit IRC | 20:02 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_keystone: Ansible 2.1.1 role testing https://review.openstack.org/353606 | 20:03 |
| cloudnull | Matias: we could do something like this http://cdn.pasteraw.com/31ujnc2eczjucsso8pv7tcx5851b9k | 20:03 |
| cloudnull | where we take the output checksum it and compair before / after ? | 20:04 |
| Matias | cloudnull: looks good | 20:04 |
| cloudnull | then we can changed_when: rc==3, failed_when: rc no in [0, 3] | 20:04 |
| Matias | but don't forget the directory may not exist | 20:04 |
| Matias | so let file module create it, or include mkdir -p in the shell task | 20:05 |
| *** TxGirlGeek has joined #openstack-ansible | 20:05 | |
| cloudnull | ideed, file module creates it then we chown it | 20:06 |
| *** catlook has quit IRC | 20:07 | |
| *** thorst_ has quit IRC | 20:08 | |
| *** TxGirlGeek has quit IRC | 20:10 | |
| *** michaelgugino has quit IRC | 20:14 | |
| *** johnmilton has quit IRC | 20:16 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_designate: Ansible 2.1.1 role testing https://review.openstack.org/353587 | 20:16 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_zaqar: Ansible 2.1.1 role testing https://review.openstack.org/353617 | 20:18 |
| *** jmckind_ has joined #openstack-ansible | 20:18 | |
| *** catlook has joined #openstack-ansible | 20:20 | |
| *** jmckind has quit IRC | 20:20 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_barbican: Ansible 2.1.1 role testing https://review.openstack.org/353567 | 20:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Add config template for qemu.conf https://review.openstack.org/366917 | 20:21 |
| *** sdake_ has joined #openstack-ansible | 20:25 | |
| *** sdake has quit IRC | 20:28 | |
| *** smatzek_ has quit IRC | 20:29 | |
| *** weezS has joined #openstack-ansible | 20:31 | |
| *** sigmavirus is now known as sigmavirus|awa | 20:38 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-tests: Ansible 2.1.1 role testing https://review.openstack.org/369038 | 20:39 |
| *** sonus has quit IRC | 20:40 | |
| *** admin0 has quit IRC | 20:41 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_zaqar: Ansible 2.1.1 role testing https://review.openstack.org/353617 | 20:44 |
| *** pcaruana has quit IRC | 20:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Make the file name for user_secrets a variable https://review.openstack.org/368156 | 20:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Newton 2016-09-12 https://review.openstack.org/367531 | 20:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 13.3.4 https://review.openstack.org/367546 | 20:47 |
| *** catlook has quit IRC | 20:48 | |
| *** itsuugo has quit IRC | 20:52 | |
| *** itsuugo has joined #openstack-ansible | 20:53 | |
| prometheanfire | odyssey4me: ? | 20:57 |
| *** itsuugo has quit IRC | 20:57 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:58 | |
| *** mgeezy has joined #openstack-ansible | 20:59 | |
| mgeezy | is anyone familiar with configuration of arp requests and arp proxy inside neutron VM's | 21:00 |
| *** itsuugo has joined #openstack-ansible | 21:00 | |
| *** thorst_ has joined #openstack-ansible | 21:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally: Remove apt-get update from Vagrantfile https://review.openstack.org/369008 | 21:05 |
| *** galstrom is now known as galstrom_zzz | 21:05 | |
| *** thorst_ has quit IRC | 21:12 | |
| *** retreved has quit IRC | 21:13 | |
| *** woodard_ has joined #openstack-ansible | 21:13 | |
| *** woodard has quit IRC | 21:16 | |
| *** woodard_ has quit IRC | 21:18 | |
| *** npawelek has quit IRC | 21:18 | |
| *** BjoernT has quit IRC | 21:21 | |
| *** askb has joined #openstack-ansible | 21:25 | |
| *** npawelek has joined #openstack-ansible | 21:26 | |
| *** itsuugo has quit IRC | 21:27 | |
| *** itsuugo has joined #openstack-ansible | 21:28 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_server: Set permissions on the openstackgit directory https://review.openstack.org/368997 | 21:30 |
| cloudnull | Matias: ^ | 21:30 |
| cloudnull | i gotta run to my next meeting. | 21:30 |
| cloudnull | bbl | 21:30 |
| *** itsuugo has quit IRC | 21:35 | |
| *** itsuugo has joined #openstack-ansible | 21:38 | |
| *** sdake_ has quit IRC | 21:38 | |
| *** sdake has joined #openstack-ansible | 21:39 | |
| *** AnarchyAo has joined #openstack-ansible | 21:41 | |
| *** npawelek has quit IRC | 21:41 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: [Docs] Add explanations about our bug triage process https://review.openstack.org/369053 | 21:41 |
| *** npawelek has joined #openstack-ansible | 21:42 | |
| *** john51 has quit IRC | 21:45 | |
| *** john51 has joined #openstack-ansible | 21:46 | |
| *** sdake_ has joined #openstack-ansible | 21:47 | |
| *** sdake has quit IRC | 21:49 | |
| *** sdake has joined #openstack-ansible | 21:50 | |
| *** sdake_ has quit IRC | 21:52 | |
| *** itsuugo has quit IRC | 21:53 | |
| *** itsuugo has joined #openstack-ansible | 21:55 | |
| *** weezS has quit IRC | 21:58 | |
| *** jperry has quit IRC | 22:01 | |
| *** KLevenstein has quit IRC | 22:02 | |
| mrda | Morning OSA | 22:03 |
| *** schwicht has quit IRC | 22:06 | |
| *** thorst_ has joined #openstack-ansible | 22:10 | |
| *** michauds has quit IRC | 22:10 | |
| hughmFLEXin | mornin mrda | 22:13 |
| mrda | o/ | 22:14 |
| *** jmckind_ has quit IRC | 22:16 | |
| *** schwicht has joined #openstack-ansible | 22:18 | |
| *** thorst_ has quit IRC | 22:18 | |
| *** mgeezy has quit IRC | 22:20 | |
| *** TxGirlGeek has quit IRC | 22:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Enable the opportunistic strategy https://review.openstack.org/349485 | 22:29 |
| *** schwicht has quit IRC | 22:30 | |
| *** schwicht_at_work has joined #openstack-ansible | 22:30 | |
| cloudnull | hows it mrda | 22:36 |
| hughmFLEXin | cloudnull: thanks for the fix, it got past that point of failure. but now another :-( | 22:38 |
| hughmFLEXin | https://gist.github.com/bbyhuy/938e695dbcda704bc8d91c950b5c41ac | 22:38 |
| hughmFLEXin | eth2.41 is fine on the node itself, so im assuming it is inside the container that is failing | 22:39 |
| cloudnull | its running the veth wiring script | 22:40 |
| cloudnull | can you run that command manually? | 22:41 |
| mrda | hey cloudnull | 22:41 |
| cloudnull | ohai mrda, how you been ? | 22:41 |
| Matias | back | 22:41 |
| Matias | cloudnull: i'll see the PR in an instant | 22:41 |
| Matias | mrda: sorry, but i need to say that your nickname sounds pretty funny in Portuguese (and I think in any language derived from Latin) | 22:42 |
| mrda | Matias: Someone has told me this before. Unfortunately it's derived from my initials. | 22:43 |
| mrda | Or perhaps it's appropriate once people see my code :) | 22:43 |
| Matias | lol | 22:44 |
| *** klamath has quit IRC | 22:44 | |
| hughmFLEXin | cloudnull: here is manual output https://gist.github.com/bbyhuy/29c6d6ef04ac6c85fbed8bbe5de6a693 | 22:46 |
| openstackgerrit | Merged openstack/openstack-ansible: Update all SHAs for 13.3.4 https://review.openstack.org/367546 | 22:46 |
| *** sdake has quit IRC | 22:48 | |
| cloudnull | hughmFLEXin: can you attach to the container | 22:48 |
| cloudnull | and run ip a l | 22:48 |
| *** openstackgerrit has quit IRC | 22:48 | |
| cloudnull | im curious if the interfaces work? | 22:49 |
| cloudnull | and are created in the container | 22:49 |
| *** openstackgerrit has joined #openstack-ansible | 22:49 | |
| cloudnull | also can you cat out /var/lib/lxc/$CONTAINER_NAME/*.ini | 22:49 |
| cloudnull | for reference | 22:49 |
| hughmFLEXin | here is ip output: https://gist.github.com/bbyhuy/0e75bc4665bfeeb4d159e38783750caf | 22:50 |
| hughmFLEXin | here is the cat: https://gist.github.com/bbyhuy/e642fafdf8b24a7accb141180fbd5607 | 22:51 |
| *** schwicht_at_work has quit IRC | 22:51 | |
| cloudnull | hughmFLEXin: in the conntainer are there network scripts? | 22:53 |
| cloudnull | anything under /etc/network/interfaces.d/ ? | 22:53 |
| *** schwicht has joined #openstack-ansible | 22:53 | |
| hughmFLEXin | eth2.41.cfg | 22:54 |
| hughmFLEXin | you said network scripts and i automatically wanted to look for centos style "network-scripts" folder...lol | 22:54 |
| cloudnull | and if you run ifup eth2.41 does it not work ? | 22:54 |
| cloudnull | hahahaha. sorry . | 22:54 |
| hughmFLEXin | RTNETLINK answers: File exists | 22:55 |
| hughmFLEXin | Failed to bring up eth2.41. | 22:55 |
| hughmFLEXin | hmmm, i notice these containers don't have vlan package installed | 22:56 |
| hughmFLEXin | or the module enabled... | 22:56 |
| *** schwicht has quit IRC | 22:56 | |
| hughmFLEXin | not sure if that is required | 22:57 |
| cloudnull | yea it may be | 22:57 |
| cloudnull | can you apt-get install vlan in the container | 22:58 |
| hughmFLEXin | yea, now it shows: | 23:01 |
| hughmFLEXin | Set name-type for VLAN subsystem. Should be visible in /proc/net/vlan/config | 23:01 |
| hughmFLEXin | RTNETLINK answers: File exists | 23:01 |
| hughmFLEXin | Failed to bring up eth2.41. | 23:01 |
| cloudnull | so if you cycle ifup/down the interface | 23:02 |
| cloudnull | do you see an appropritate active interface? | 23:02 |
| *** hblah has joined #openstack-ansible | 23:03 | |
| *** itsuugo has quit IRC | 23:04 | |
| *** hblah has quit IRC | 23:05 | |
| hughmFLEXin | it shows up in ifconfig | 23:06 |
| hughmFLEXin | but stderr still throws failure | 23:06 |
| *** itsuugo has joined #openstack-ansible | 23:06 | |
| cloudnull | is the vlan module installed / loaded on the host? | 23:06 |
| hughmFLEXin | yup... | 23:06 |
| hughmFLEXin | from controller to controller i can ping the vlans to each other | 23:07 |
| cloudnull | what does the interface file look like within the container? | 23:07 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Remove assumption that the neutron_lbaas var is set https://review.openstack.org/368905 | 23:09 |
| hughmFLEXin | cloudnull: https://gist.github.com/bbyhuy/4b19c1b0e4ac2c621a9d2d257dd8ac20 | 23:09 |
| *** hblah has joined #openstack-ansible | 23:10 | |
| cloudnull | that all looks right | 23:11 |
| hughmFLEXin | cloudnull: would it help if we setup a webex and i can give you remote access? | 23:12 |
| cloudnull | i wish , im on a linux system , webex is a no go :) | 23:13 |
| *** klamath_ has quit IRC | 23:13 | |
| hughmFLEXin | hah.. | 23:14 |
| cloudnull | when you ifup eth2.42 you get an address according to ifconfig ? | 23:14 |
| cloudnull | and | 23:15 |
| cloudnull | you see trraffic in and out ? | 23:15 |
| *** thorst_ has joined #openstack-ansible | 23:15 | |
| hughmFLEXin | eth2.42? | 23:15 |
| hughmFLEXin | or eth2.41 | 23:15 |
| cloudnull | yes | 23:15 |
| cloudnull | or | 23:15 |
| cloudnull | whatever it is | 23:16 |
| cloudnull | sorry i typo'd it | 23:16 |
| hughmFLEXin | yea ifconfig shows the address and there is tiny bits of rx/tx | 23:16 |
| hughmFLEXin | RX bytes:110350 (110.3 KB) TX bytes:16184 (16.1 KB) | 23:16 |
| cloudnull | is there anything else on that network ? | 23:16 |
| cloudnull | something you can ping? | 23:16 |
| hughmFLEXin | nope....i actually don't know where that subnet came from 172.29.239.x | 23:18 |
| cloudnull | that's set in /etc/openstack_deploy/openstack_user_config.yml | 23:19 |
| *** schwicht has joined #openstack-ansible | 23:20 | |
| hughmFLEXin | im not using 239 anywhere, I use 236/240/244 | 23:20 |
| cloudnull | are they /22 ? | 23:20 |
| hughmFLEXin | yes.... | 23:20 |
| cloudnull | 172.29.236.0/22 last ip is 172.29.239.255 | 23:21 |
| hughmFLEXin | ah got it | 23:22 |
| hughmFLEXin | so from the container i can actually ping the hosts through that subnet with their respective vlan ips | 23:23 |
| *** thorst_ has quit IRC | 23:23 | |
| cloudnull | ok so thats good. | 23:23 |
| cloudnull | so when you run the veth wiring script is it still exiting somthing not 0 or 3? | 23:24 |
| *** weezS has joined #openstack-ansible | 23:25 | |
| *** weezS has quit IRC | 23:25 | |
| hughmFLEXin | return 1 | 23:26 |
| hughmFLEXin | Failed to bring up eth2.41. | 23:26 |
| hughmFLEXin | root@osa-controller-3:~# echo $? | 23:26 |
| hughmFLEXin | 1 | 23:26 |
| cloudnull | do you think its failing something in the config file ? | 23:27 |
| cloudnull | maybe sysctl -w net.ipv4.conf.$IFACE.arp_notify=1 | 23:28 |
| hughmFLEXin | hm the wiring script also has this in there somewhere | 23:28 |
| hughmFLEXin | Device "0ea8d9d5" does not exist. | 23:28 |
| cloudnull | or ip link set $IFACE address $(cat /sys/class/net/$IFACE/address) ? | 23:28 |
| cloudnull | hum. | 23:28 |
| cloudnull | just for shits and grins, can you lxc-stop -kn $name; lxc-start -dn $name; | 23:31 |
| cloudnull | and see if things are the same? | 23:31 |
| hughmFLEXin | yup same....would've been nice if a cycle fixed it lol | 23:32 |
| cloudnull | it wouldve | 23:33 |
| cloudnull | i'm wondering if its the post up entries that are failing | 23:33 |
| cloudnull | seen as it's getting an address and such | 23:33 |
| *** itsuugo has quit IRC | 23:34 | |
| hughmFLEXin | ah | 23:36 |
| hughmFLEXin | might be | 23:36 |
| *** itsuugo has joined #openstack-ansible | 23:36 | |
| hughmFLEXin | so running | 23:36 |
| hughmFLEXin | ysctl -w net.ipv4.conf.eth2.41.arp_notify=1 | 23:36 |
| hughmFLEXin | takes me to the incorrect directory | 23:37 |
| hughmFLEXin | .../../ipv4/conf/eth2/41/arp_notify: No such file or directory | 23:37 |
| hughmFLEXin | when correct directory is | 23:37 |
| hughmFLEXin | .../../ipv4/conf/eth2.41/arp_notify | 23:37 |
| *** itsuugo has quit IRC | 23:41 | |
| openstackgerrit | Merged openstack/openstack-ansible: Update all SHAs for Newton 2016-09-12 https://review.openstack.org/367531 | 23:42 |
| *** itsuugo has joined #openstack-ansible | 23:42 | |
| *** Jeffrey4l has quit IRC | 23:43 | |
| *** Jeffrey4l has joined #openstack-ansible | 23:43 | |
| *** itsuugo has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!